From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-0.9 required=5.0 tests=BAYES_00,FORGED_GMAIL_RCVD,
	FREEMAIL_FROM autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: 103376,c4e4e40240a86624
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news1.google.com!postnews.google.com!s18g2000prg.googlegroups.com!not-for-mail
From: Emile8 <pocchiola.yves@gmail.com>
Newsgroups: comp.lang.ada
Subject: Re: Anyone see a problem in 5.7 ncurses Ada?
Date: Wed, 9 Mar 2011 12:28:53 -0800 (PST)
Organization: http://groups.google.com
Message-ID: 
 <431bc7ac-07d3-4ccc-abb6-18091cb5a0db@s18g2000prg.googlegroups.com>
References: <slrninfj90.kmk.localhost@example.org>
NNTP-Posting-Host: 90.27.116.165
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
X-Trace: posting.google.com 1299702533 14236 127.0.0.1 (9 Mar 2011 20:28:53
 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Wed, 9 Mar 2011 20:28:53 +0000 (UTC)
Complaints-To: groups-abuse@google.com
Injection-Info: s18g2000prg.googlegroups.com; posting-host=90.27.116.165;
 posting-account=u9P-fgoAAAB1qVhsrmxVpcdi-eIb_an8
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/5.0 (X11; U; Linux x86_64; fr; rv:1.9.2.15)
 Gecko/20110303 Ubuntu/10.10 (maverick) Firefox/3.6.15,gzip(gfe)
Xref: g2news1.google.com comp.lang.ada:18009
Date: 2011-03-09T12:28:53-08:00
List-Id: <comp.lang.ada>

On 9 mar, 19:55, localh...@example.org wrote:
> I had built some programs against an earlier version of ncurses, I can't
> remember whether it was 5.5 or later and they were working fine. When I
> rebuilt them against 5.7 a few of them broke. I have tracked it down to
> incorrect values being returned by get_size. For a 24x80 xterm get_size is
> returning 25 columns and 81 lines. This makes the values actually 2 higher
> than they should be, because the numbers are supposed to be
> zero-relative. That is the lines should be 23 and the columns 79. Has anyone
> else noticed this? I don't know whether it's in the Ada binding or in
> ncurses itself. I would think if ncurses itself was broken people would have
> noted and fixed it but then again I don't think the Ada binding has been
> touched in years (and wow, it's good).

I confirm your observation on the Ada binding of the version 5.7 of
ncurses. The numbers of lines and columns are 1 higher than the actual
ones. I don't know if this is relevant but Get_Size gives the number
of lines and columns of a window in the terminal. It seems so that the
standard_window is a bit greater than its terminal.

with Terminal_Interface.Curses; use Terminal_Interface.Curses;
with ncurses2.util;             use ncurses2.util;
with ada.text_io; use ada.text_io;
with Ada.integer_Text_IO; use Ada.integer_Text_IO;

procedure test1 is
   NL : Line_count;
   NC : Column_count;
begin
   init_screen;
   Get_size (number_of_columns => NC, number_of_lines => NL);
   put(integer(NL));
   put_line("");
   put(integer(NC));
   delay 10.0;
end test1;


Here after is the source code of Get_Size. It could be that the
problem comes from the value of Offset_XY.

procedure Get_Size
     (Win               : in Window := Standard_Window;
      Number_Of_Lines   : out Line_Count;
      Number_Of_Columns : out Column_Count)
   is
      function GetMaxY (W : Window) return C_Int;
      pragma Import (C, GetMaxY, "getmaxy");

      function GetMaxX (W : Window) return C_Int;
      pragma Import (C, GetMaxX, "getmaxx");

      Y : constant C_Int := GetMaxY (Win)
                          + C_Int (Offset_XY);
      X : constant C_Int := GetMaxX (Win)
                          + C_Int (Offset_XY);
   begin
      Number_Of_Lines   := Line_Count (Y);
      Number_Of_Columns := Column_Count (X);
   end Get_Size;